VAELLA
VAELLA provides a privacy-first Vision AI platform that gives building operations teams real-time awareness of people flow in crowded spaces like metro stations, airports, and office buildings. The system combines ceiling-mounted laser depth sensors with a digital twin to automate responses such as elevator rerouting, security dispatch, and maintenance ticketing, while remaining physically incapable of capturing biometric data.
- Artificial Intelligence
- Internet of Things
- Property Technology
- AI Agents
- Hardware
- Software Only
- Data & Analytics
Funding
Founders
Founder details are not available yet.
Product
Problem
Building operations teams struggle to manage crowded spaces like metro stations, airports, and office lobbies because existing systems only capture equipment telemetry, badge swipes, and ticket counts—not what people are actually doing in real time. Traditional camera-based analytics raise privacy concerns and fail in dense crowds, while disconnected dashboards leave operators reacting to disruptions after they occur rather than preventing them.
Solution
Vaella provides an AI operator platform that combines ceiling-mounted vision sensors with a digital twin to give buildings real-time awareness of people flow. The vision sensors use laser-based depth sensing to detect occupancy, density, flow direction, queues, and dwell times—producing structured event data without capturing images, faces, or biometric information. The digital twin stores spatial geometry, live state, and historical records across building systems, enabling the AI operator to compare current events against past patterns and determine whether situations are routine or unusual. The platform can automatically reroute elevators, dispatch security, file maintenance tickets, and send contextual alerts to team chat tools, while also recommending operational changes backed by simulation.
Target Audience
Primary customers are operations teams at metro stations, airports, cruise terminals, and large office buildings that need real-time crowd awareness and automated responses to maintain service continuity during disruptions.
Features
- Privacy-by-physics vision sensors using laser depth sensing that are physically incapable of producing biometric identification data, with no lens, pixels, or image capture
- On-device processing where raw distance measurements never leave the unit; only anonymous tracks, zone counts, density readings, and flow vectors are emitted
- Digital twin combining spatial geometry, live state, and queryable historical archive—including sensor events, commands, overrides, and rule changes
- Automated elevator control that pre-positions cars from upstream flow data before queues form and redirects traffic when escalators fault
- Plain-language chat interface with contextual alerts, daily operation digests, weekly equipment reliability summaries, and monthly utilization reports
- Simulation engine that runs failure scenarios and operational changes against the building's own historical data to project outcomes before implementation
- 3D spatial view accessed via deep-links from alerts for rapid incident investigation and contractor briefings
- Integrations with building management systems, access control, maintenance ticketing systems, and team chat platforms